Our overall aim is to analyze intergenerational exchanges in the context of specific kin groups that include persons aged 65 and over. Exchanges to and from the elderly in the currencies of time, space, and money will be considered. The requisite data for such analyses includes information on the relevant statuses of the reference person (or ego) and all kin of that ego, the timing of significant transitions in the life-cycles of ego and kin, and on the direction, timing and types of flows across generations. Because no single data set provides adequate coverage of all these domains, we propose a two-part research strategy that (1) exploits several existing data sources that contain information on the structure and dynamics of kin networks, and (2) integrates the results of such analyses by means of a microsimulation model of kinship, KINSIM, previously developed by the applicant. In particular, we propose a four- year effort to: 1. analyze existing surveys, by means of various statistical models (primarily varieties of multinomial logit specifications), in order to estimate both cross-sectional and panel models of intergenerational exchanges as behavioral outcomes of kin structures and other hypothesized factors, including race; 2. extend and refine KINSIM by introducing (i) interdependencies that arise from the intergenerational transmission of fertility patterns; and (ii) transitions to and from institutions; 3. use the simulated kin structures as a context for the estimated models of intergenerational exchanges--that is, to embed models of the consequences of kin network structures in a model of the structures themselves. In this way we will be able to calculate life-cycle patterns of intergenerational exchanges, and determine the incidence, duration, timing, and net lifetime volume of various exchanges; this would otherwise require a life-time of data collection. Results from the statistical models that reference the availability of kin and their attributes will enable us to address, among others, questions regarding the substitutability of different types of kin and transfer arrangements; results from the integrated simulation exercised will address questions pertaining to the intergenerational transfer burden across the full population of kinsmen.